Abstract
This paper provides insights into a failed attempt at participatory video (PV). PV has long been favoured by researchers working with marginalised communities. However, there is limited discourse about when the method is and is not appropriate, and few published examples of when it has failed. It is important to critique research methods, and for researchers to be transparent about when research is not carried out as originally intended. Such reflection allows us to refine the methods we use and improve our research. This paper explores what a failed PV project with Takatāpui/LGBTIQ+ people who had experienced homelessness taught us about the stigmatised nature of both homelessness and LGBTQ+ identities. Furthermore, it shows how methods that do not allow for participants to maintain their anonymity are sometimes not the right choice when researching stigmatised issues.
Keywords
Introduction
The aim of this paper is to describe the learnings from a failed attempt at participatory video (PV). PV repositions research participants as co-producers of knowledge, in partnership with the researcher (Gubrium et al., 2014). This aims to lessen the power imbalances between researcher and participant, and to centre the voices of marginalised groups who might not otherwise be heard. PV is most frequently used within marginalised communities, as an effective way of including vulnerable groups in order to produce more equitable outcomes (Haynes & Tanner, 2015). The methodology is said to empower participants, and result in knowledge produced by and for communities (Haynes & Tanner, 2015). However, this was not achieved in the context of this project. Our PV project sought, but did not succeed, to explore experiences of Takatāpui/LGBTIQ+ homelessness.
Participatory research methods, particularly PV, have been used across a range of disciplines; including geography, social work, public health, education, anthropology, medicine, and management studies (Gubrium et al., 2014; Holland et al., 2010; Jarldorn, 2019; Kindon, 2003, 2016b; Kingery et al., 2016; Prins, 2010; Sitter, 2015; Walsh, 2016; Whiting et al., 2016; Zoettl, 2013). Broadly speaking, participatory research aims to create social change, and gives space for participants to engage in all aspects of the research process, allowing them to generate knowledge in collaboration with researchers (Kindon et al., 2007). Our choice to attempt PV in this project drew from Community-Based Participatory Research (CBPR). CBPR is a collaborative approach to research that involves members of the community who are the focus of the research, organisational representatives, and researchers (Israel et al., 1998). Clark and Ventres (2016) note that CBPR is both a process and product, ongoing over time, between researcher and community members. We drew inspiration from CBPR methodologists such as Barbara Israel, Eng, Schulz, and Parker (2013; 1998, 2013). 1 PV can also be classed as a visual arts-based method. Rice and Mündel (2018) explain that arts-informed research has the power to open up conversations about difficult or sensitive topics in a range of settings. Arts-based research is increasingly being used across a range of disciplines (Leavy, 2008). In terms of the validity of such methods, Leavy (2008) explains that artistic methods ought to move away from “rigour” to “vigour.” That is, evaluation of such methods is not a uniform approach, rather, their so-called successes are inherently linked to the research objectives (Leavy, 2008). This paper is an effort to increase the vigour of this failed project; to not shy away from our failings, but to embrace them and acknowledge what they have taught us.
PV has been used by researchers in a wide range of disciplines. Our PV project drew on the work of Shaw and Robertson (1997; 2012, 2015, 2016) and Sitter (2012, 2015), who outline example processes and fundamental ground rules to PV projects. We present further detail on these processes in the methods section. Alongside this, we viewed a range of YouTube videos that were either about the PV process itself, or which were video outputs from PV projects (A Participatory Video Made by Chivoko Village, Solomon Islands - YouTube, 2011; Participatory video: Curse of the Monsoon - YouTube, 2016; Baumhardt, 2011; Morales, 2015). The videos we watched were from the geography and environmental science disciplines, with a focus on education, climate change, and empowerment. Much of what we saw, and read, about PV focused on supporting participants to express their needs in relation to a specific issue; be that education for women, the effect of climate change, disability rights, gender inequities, or mental illness, to name a few (Singh et al., 2018; Sitter, 2015; Whitley et al., 2020). We felt that the use of PV in our own project would enable us to support participants to express what their needs were (particularly in relation to housing support services) during their periods of homelessness. Our goal was, if all participants consented, to show the resultant videos to housing support organisations so they could better account for the needs of people who identify as Takatāpui/LGBTIQ+.
As already noted, we failed to carry out our PV project, and instead worked towards this goal through other research methods. However, we were determined that the lessons learned in our attempt in using PV should not be lost. Sousa and Clark (2019) propose that the invisibility of research failures leads to underestimations of how frequently researchers fail, leading those of us who do fail in our research to question our abilities. There is a slowly-growing body of evidence regarding research failure (Cameron, 2007; CohenMiller et al., 2020; Frazier, 2020; Nairn et al., 2005; Prior, 2014; Rice & Mündel, 2019; Ross & Call-Cummings, 2020; Sousa & Clark, 2019). Prior (2014) reminds us that research, particularly research based on qualitative methods such as interviews, is a jointly constructed activity between researcher(s) and participant(s). This, he explains, forces us to challenge the roles and procedural norms within our chosen methods (Prior, 2014). Doing so allows us to reframe failure as something to learn from. In her discussion of a failed photo-elicitation method with resettled refugees, Frazier (2020) illustrates a number of reasons why participatory research can fail: limited time in the field causes pressure to recruit and conduct the research in a timely fashion, participant burden, power dynamics, and affiliations with outside organisations (specifically, both actual affiliations hindering participation, and perceived affiliation encouraging participation for unwanted reasons). Furthermore, this literature notes the importance of discussing failure as a practice of reflexivity (Frazier, 2020; Ross & Call-Cummings, 2020; Sousa & Clark, 2019). Doing so strengthens the research methods we employ, encourages us to think critically about how we research, and the effect our research has on our participants. Discussing such failure reflects the complexity of employing qualitative methods, and of the communities, people, and phenomena we seek to research (Sousa & Clark, 2019). This paper provides a modest contribution to such a task.
For this project, we sought to use PV to explore Takatāpui/LGBTIQ+ people’s experiences of homelessness in Aotearoa New Zealand (henceforth referred to as Aotearoa). The failed PV project we discuss in this paper fit into BF’s wider doctoral project, which was one of the first of its kind to explore Takatāpui/LGBTIQ+ homelessness in Aotearoa. The acronym LGBTIQ+ stands for lesbian, gay, bisexual, transgender, intersex, queer, and other minority gender and sexual orientation identities (such as pansexual, non-binary, and asexual). The inclusion of the + sign is intended to be inclusive of the additional identities that are not included within the LGBTIQ acronym. The word takatāpui has been translated to mean “intimate partner of the same sex,” and in the 1980s it was gifted to Māori 2 LGBTIQ+ people by academic and activist Ngahuia Te Awekotuku (Kerekere, 2017). The word is now widely used among LGBTIQ+ identifying Māori as both an identity in and of itself, and as an umbrella term (Kerekere, 2017).
International research shows that 20–40% of homeless populations identify as LGBTIQ+ (Ecker, 2016; Ecker et al., 2018; Fraser et al., 2019). LGBTIQ+ people who experience homelessness face increased risk of poor outcomes such as mental ill health, HIV, substance use, sexual assault, and discrimination (Dolamore & Naylor, 2018; Drescher et al., 2019; Robinson, 2020; Tyler & Schmitz, 2018; Whitbeck et al., 2004). Furthermore, LGBTIQ+ people who experience homelessness face high levels of discrimination and stigma due to both experiencing homelessness and their LGBTIQ+ identities (Bardwell, 2019; Ecker et al., 2019; Gattis, 2013; Kidd, 2007; Robinson, 2020; Shelton, DeChants, et al., 2018a; Shelton, Wagaman, et al., 2018b). We have previously advocated for an intersectional systems-thinking approach to LGBTIQ+ homelessness (Fraser et al., 2019, 2021). Such an approach ensures that LGBTIQ+ homelessness is not individualised, but, rather, that our analyses consider structural issues (Fraser et al., 2019, 2021).
In the Aotearoa context, emerging research on transgender and gender diverse homelessness by Vandenburg et al. (2021) utilised photovoice methods. Early findings highlight the importance community engagement, insider research, and scholar-activism when conducting such research (Vandenburg et al., 2021). Our research identifies commonality in Takatāpui/LGBTIQ+ experiences prior to homelessness, including: the pervasiveness of instability (especially in regards to family relationships, finances, and housing), having to grow up fast due to social and material conditions, experiences of looking for housing in stressed markets, and systems failures that resulted in a lack of autonomy (Fraser et al., 2021). Our work has shown that the majority of participants had experienced stigma due to both their LGBTIQ+ identities and their experiences of homelessness (Fraser et al., 2021). Furthermore, many participants reported still feeling considerable levels of shame about having experienced homelessness, due to the stigmatised nature of homelessness (Fraser et al., 2021). Shame puts the onus of homelessness onto the individual; it shames them and their project of the self, instead of shaming the multiple systems that have failed them (Farrugia, 2011a, 2011b). Instead of positioning the homeless individual as a “failure,” or someone who diverges from social norms, and instead focussing on the failed systems that stigmatise and shame people for experiencing homelessness, we can better work to address Takatāpui/LGBTIQ+ homelessness.
The intention behind the PV project, and the primary reason for why it was chosen, was to ensure that LGBTIQ+ homelessness scholarship in Aotearoa began from a place which centred the voices of those with lived experience. Additionally, if the project has succeeded, and if all participants involved consented, the video output would have been shared with service providers and the wider public in order to initiate public discussions about the specific needs of Takatāpui/LGBTIQ+ people who are, or have been, homeless in Aotearoa. This paper explores the methods used to conduct PV, the challenges faced, and the lessons learned. The difficulties encountered in this project provide valuable insights into why PV may not be suitable for highly stigmatised populations such as Takatāpui/LGBTIQ+ people who have experienced homelessness. The key findings presented in this paper provide an example of how researchers can critically discuss the limitations of a research methodology. Furthermore, the findings presented show that there are difficulties in engaging economically precarious and highly stigmatised research participants in participatory methodologies.
Methods
As mentioned, the PV project formed part of BF’s doctorate on Takatāpui/LGBTIQ+ homelessness. The aims of BF’s thesis were: (1) To understand the viewpoints and perspectives of Takatāpui/LGBTIQ+ people who are, or have been, homeless. (2) To understand how both government and wider support systems shape the experience of Takatāpui/LGBTIQ+ homelessness. (3) To further the knowledge base of homelessness in Aotearoa through high quality information.
This thesis was some of the first research on Takatāpui/LGBTIQ+ homelessness in Aotearoa. Ethics approval for both the interviews and PV project was granted by the University of Otago’s Human Ethics Committee, reference 18/147. Semi-structured interviews with Takatāpui/LGBTIQ+ identifying people who had been, or were, homeless were conducted between October 2018 and February 2019. Our inclusion criteria was anyone who identified as Takatāpui/LGBTIQ+, met the Aotearoa definition of homelessness (Amore et al., 2013), and were located in Te Whanganui-a-Tara (Wellington) so they could be interviewed in person. There were eight participants who met the Aotearoa definition of homelessness; we aimed to recruit these participants for the PV project.
Recruitment and Participants
Participant recruitment was carried out via posters in key locations across Te Whanganui-a-Tara, emails and social media, word of mouth, and researcher visits to additional key locations such as The Free Store 3 and Tīwhanawhana 4 to talk to potential interviewees. Some key locations—such as homeless support organisations—were not visited. This was a conscious decision to seek out the viewpoints and experiences of people who might not have utilised housing service providers. BF was hesitant to align themself with service providers as they knew, from experience volunteering within the Takatāpui/LGBTIQ+ community for a number of years, that there is limited trust of organisations meant to provide support. The entire team agreed on this approach. This research utilised the critical paradigm, which allows for the researcher’s values to be “central to the task, purpose, and methods of the research” (Ponterotto, 2005, p. 129). This gave space for BF’s insider position as a member of the Takatāpui/LGBTIQ+ community and the values they attach to this, which enabled them to understand the research findings on a level that an outsider to the Takatāpui/LGBTIQ+ community might not have.
Demographics.
BF intended to carry out PV with the majority of the eight participants they interviewed for their wider doctoral project. Two of the initial interviewees said that they did not want to participate in the PV sessions; the other interviewees expressed interest in participating in the sessions. We spent a considerable amount of time attempting to schedule the sessions at a time that best suited all participants (discussed below). However, of those remaining six participants, only two (Avery and Omar) attended the first video session, and only one of those two (Avery) attended the second session; PV sessions require the involvements of a group of people (Shaw, 2016). Two PV sessions were held in mid-2019. BF wrote memos after each session to reflect on the findings and challenges.
Data Collection
PV serves to strengthen community ties and to amplify the voices of marginalised communities. This methodology also had the opportunity to strengthen ties between different communities, by bringing together Takatāpui/LGBTIQ+ people with experiences of homelessness and the organisations that were a part of these experiences, such as housing service providers. Green and Thorogood (2014) suggest the purpose of participatory research should be to engage in dialogue with marginalised people in order to further emancipation or critical awareness. This research sought to raise awareness and consciousness of the challenges Takatāpui/LGBTIQ+ people experiencing homelessness can face when accessing homelessness services—both as Takatāpui/LGBTIQ+ people, and as people who have experienced homelessness. The intention of using PV in this project was to enable participants to co-produce knowledge about their needs and experiences of accessing homelessness service providers, so that the resulting videos could be shared with service providers in order to raise awareness of Takatāpui/LGBTIQ+ homelessness and the specific needs of this community.
PV repositions research participants as co-producers of knowledge who are in partnership with the researcher (Gubrium et al., 2014). The method is an effective means to include and work with marginalised groups to produce more equitable outcomes (Haynes & Tanner, 2015). It is often preferred as it results in knowledge produced by and for participants, as well as their communities (Haynes & Tanner, 2015). The aim of this approach is to lessen the power imbalances between researcher and participant, and to amplify marginalised groups who might not otherwise be heard by those in power. The core assumption of PV as a methodology is that community members are experts in their own situations, and the role of the community should involve active participation (Sitter, 2012). The methodology provides a means of both recording and developing ideas to those who might not be familiar—or at ease with—writing about complex issues and allows the group to share their opinions in a way that goes beyond what other methodologies such as interviews and focus groups are capable of. Sitter (2012) notes PV can raise questions as to why certain histories are valued more than others. PV has the potential to challenge spaces where, historically, for example, people who have not experienced homelessness speak on behalf of people who have experienced homelessness. It also has the potential to challenge spaces where non-LGBTIQ+ people speak on behalf of LGBTIQ+ people.
The PV sessions were each several hours long; participants were given koha 8 for their time and tea, coffee, snacks, and lunch were provided. 9 The atmosphere of the sessions was relaxed, and aimed at facilitating discussion and giving participants the tools to design and create a video about what was important to them. The sessions were primarily based on the work of Shaw and Robertson (1997; 2012, 2016) who provides four ground rules for PV projects. These are; all participants take turns both operating equipment and assuming production roles—roles are rotated every exercise/shot; all participants perform in front of the camera each session; all video material is played back on a monitor after each recording; video recordings are confidential and only shown externally if all participants decide and agree (Shaw, 2012). Furthermore, she provides basic exercises practitioners are able to use as a guideline in their own work.(Shaw, 2012). The procedure of this exercise is for one participant to act as camera operator, whilst the other participants face the camera (Shaw, 2012). Recording starts, and one-by-one participants make a statement on a pre-arranged topic and recording stops when everyone has contributed (Shaw, 2012). Participants then watch the recording and discuss it. The purpose of this exercise is to encourage self-expression, create space for diversity of opinion, and ensure all participants are able to both speak and be heard (Shaw, 2012). The content generated in such an exercise can focus on a number of things such as experiences, perceptions, opinions, and/or evaluations (Shaw, 2012). This example was used to make a runsheet for the sessions, which used data from BF’s wider PhD findings as prompts for discussion. 10 During the second session, when there was only one participant, the runsheet was adapted and BF and their primary supervisor facilitated and engaged in a discussion with the participant about their experiences of service providers. The sessions used a Panasonic HC-V800 camcorder and RODELink Filmmaker Kit microphones.
Data Analysis
Due to the failed nature of this project, we did not have any video outputs which were able to be shared, or formally analysed. Additionally, due to ethics requirement and not obtaining consent from each participant, we are unable to share the footage we did gather. However, the footage and brainstorming notes from the first PV session contained some useable data. While the data was not formally coded, BF re-watched the footage and took notes on the content, writing memos about it after viewing. This, alongside the brainstorming notes, was used to identify three key categories; bullying, trust, and intersecting identities. As discussed above, findings (including emergent data categories) from the semi-structured interviews BF conducted prior to the interviews were used to prompt discussions during the sessions.
Results
Conducting PV brings participants together to discuss their shared experiences and tell their stories in their own words. This project aimed to have as many interview participants as possible take part in the PV sessions. However, this did not happen as anticipated. Instead, only two participants attended the first session, and only one attended the second session. This section describes the process and results of organising the PV sessions and the challenges that arose. Specifically, we focus on the scheduling, recruitment and visibility, and cancellation of the PV project. This section then ends with an overview of the results from the workshops that we ran before deciding to cancel the PV project. While this specific project failed, it has much it can teach us about participant engagement and working with stigmatised communities. Researchers in Aotearoa wanting to explore Takatāpui/LGBTIQ+ homelessness through participatory methods would do well to consider alternative methods such as photovoice, until such a time as when there is increased public awareness about the issue and willingness to publicly discuss experiences of Takatāpui/LGBTIQ+ homelessness.
Scheduling
Scheduling the PV session proved to be difficult and took much more time than anticipated. Overall, it took nearly 5 months of scheduling attempts to find a suitable date and time. Scheduling was done via email, and eventually, anonymous Doodle polls. 11 This gave participants a range of dates and times to choose from in order to ensure as many as possible were able to attend.
We initially scheduled the first session for mid-March 2019 (Figure 1). However, many of the participants were involved in organising and attending the Te-Whanganui-a-Tara pride festival, as well as the International Lesbian, Gay, Bisexual, Trans and Intersex Association (ILGA) World Conference 2019, which were both happening at the end of March. After BF consulted both their supervisory team and the participants, we scheduled the first video session for the 13th of April. On the 11th of April, one participant said they would be unable to make this specific session. The following day, another participant dropped out of the video project. With only one participant confirmed, we decided to postpone the session. However, the one confirmed participant informed BF he would be away for the following month. As he had been the most consistent participant, we decided to postpone the video sessions until he was available at the end of May. We began consulting participants again in mid-May, and we settled on the 29th of June as a date that the four remaining participants would be available. We felt the PV would benefit from a greater number of attendees, even if BF had not interviewed them for their PhD. Additional recruitment of new participants was attempted through social media and contacting key people within the Takatāpui/LGBTIQ+ community in The-Whanganui-a-Tara; however, it was not successful. There was limited community interest in the video project. Timeline of scheduling the participatory.
Part of the difficulty in scheduling the sessions can be attributed to the specific demographic this research looks at. Cornwall and Jewkes (1995) explain that time can be a significant barrier in participatory research. Firstly, participation is time-consuming, and secondly, potential participants are busy trying to secure the basic necessities of life (Cornwall & Jewkes, 1995). The majority of participants in this project were still living precarious lives and struggling to find financial and housing stability. This particular cohort of people all had varying schedules; some were unemployed and receiving a government benefit, some worked full-time jobs during standard business hours, some worked in the hospitality industry during the evenings, and some worked multiple part-time jobs. They also balanced this with volunteering, community, and family commitments.
Recruitment and Visibility
Community uptake and additional recruitment proved to be a challenge in conducting PV, primarily due to the lack of anonymity this method affords participants. Responses from the Takatāpui/LGBTIQ+ community had been overwhelmingly positive when BF was recruiting participants to interview. However, when BF attempted to do additional recruitment for the video project, there was limited enthusiasm from the individuals and Takatāpui/LGBTIQ+ organisations BF had reached out to. The limited enthusiasm could, in large part, be due to the differences in visibility between the two methods. One-on-one interviews enabled participants to remain anonymous, allowing them to keep their Takatāpui/LGBTIQ+ identities and experiences of homelessness private. PV did not allow this. The research design and ethical considerations had stipulated that the videos would not be viewed by anyone outside of the research team unless all PV participants agreed to wider distribution. However, this did not allow for full anonymity. Participants would have had to meet each other, and, to an extent, share their stories during the video sessions.
Two participants explicitly stated their discomfort with the PV project; from the outset of our interview Ayeisha was clear that she did not want to partake in the PV project as she wanted to remain anonymous. Additionally, when BF was scheduling the PV sessions, Thom, who had initially indicated interest in the project, emailed BF to withdraw from the project because our interview had been the first time he had discussed his experiences of homelessness and it had resulted in a negative sense of self in regards to his behaviours during these periods of his life. We respected his wish to not take part in the PV project. We emphasised that the research team viewed his actions as displaying considerable strength of character and ability to survive difficult situations, and directed him to support organisations if he felt the need for ongoing support. This withdrawal from the PV project evidences both the amount of shame and stigma participants experienced, but also the lack of full anonymity inherent in the method. Even if the PV process had been altered so as to ensure participants did not have to appear in front of the camera, they would have still have had to come together with the knowledge that they were all members of the Takatāpui/LGBTIQ+ community who had experienced homelessness. Additionally, this was a key reason as to why we did not simply pivot to a different approach, such as photo-elicitation methods. We felt that the stigma our participants had experienced, as expressed in their interviews, was so great that using any method which required group interaction and visibility was asking too much of our participants.
The lack of visibility, and ongoing stigmatisation, of Takatāpui/LGBTIQ+ homelessness was also reflected in a memo that BF wrote on the 20th of June, 2019, which said: “Homelessness and LGBTIQ+ [identities] are very stigmatised and if no-one is really talking about this publicly yet, it’s hard and scary to agree to…open yourself up that much. PV is well-intentioned but…it doesn’t feel like the right fit for where the community is currently at…It’s feeling very forced and difficult…It doesn’t feel participatory or like a partnership at all. It feels like academics have chosen and stuck with a method that the community doesn’t want.”
This memo highlights how participating in an intensive research method about a deeply personal topic made people feel vulnerable, and gives insight into the lack of uptake for this project. Furthermore, this memo also highlights that the project was not participatory in the way the method had been chosen; while the sessions themselves had been participatory, the PV project as a whole had not been.
Cancellation
As mentioned, the first session only had two participants, and the second session only had one participant. During the second session, Avery, the sole participant, remarked that it did not feel very participatory to her because there were no other participants. A memo BF wrote after the session reflected on this:
“I was in two minds about how I should react to this because on the one hand I totally agree with what she was saying but I guess we [my primary supervisor and I] did not want to admit that the whole exercise had kind of been a failure… we just kind of made a comment about how next time we were definitely going to get more participants.”
Avery said that it did not seem like people wanted to talk about these two difficult issues on camera (both homelessness and Takatāpui/LGBTIQ+ identity). Avery was right: the video sessions were not the group activity required by the method. Due to our inability recruit and schedule sufficient numbers of participants we decided to cancel the rest of the project. We reflected that consulting with the community on appropriate methods prior to beginning our project would have made success more likely
PV Session Findings
In spite of the many difficulties conducting this phase of the research, there were some valuable findings from the two sessions. While no actual PV video was made, we had footage from the sessions, particularly the first, in which our two participants responded to, and expanded on, our prompts in the runsheet. These findings fell into the categories of bullying, trust, and intersecting identities. The purpose of the PV project was to explore participants’ experiences—or lack thereof—of service provision and pathways through services, so that the resultant videos could be shown to service providers.
Bullying, particularly within housing service organisations, was a significant concern for both Avery and Omar. They noted that bullying can happen to people who are visibly Takatāpui/LGBTIQ+, or even just people who “look gay.” A notable element of Takatāpui/LGBTIQ+ homelessness is shelter inaccessibility (Abramovich, 2016; Coolhart & Brown, 2017; Maccio & Ferguson, 2016; Piche et al., 2018). Part of this is the bullying and discriminatory attitudes of other service users. Bullying, specifically that related to their Takatāpui/LGBTIQ+ identities, within service organisations can make Takatāpui/LGBTIQ+ clients feel particularly vulnerable and unsafe, as well as reluctant to continue receiving support from such organisations (Abramovich, 2016; Coolhart & Brown, 2017; Maccio & Ferguson, 2016; Piche et al., 2018). This was reflected in the PV sessions, with Avery and Omar both noting that bullying made them less likely to engage with services again, even if they needed the help that the organisations were providing.
Trust, or the lack thereof, is an important aspect of how an individual interacts with a service provider. Avery discussed how if a transgender person had a negative experience with a medical professional or at Work and Income New Zealand (WINZ), 12 then they would be less likely to trust other service providers meant to support them. Within Takatāpui/LGBTIQ+ and homeless communities, trust must be earned. Avery did not automatically trust service providers, as she had experienced many negative interactions with people meant to support her, including some directly relating to her gender identity (such as trying to receive government support for transition-related costs). When discussing how this trust could be built, Avery noted that it requires more than an organisation simply hanging a rainbow flag in a prominent place. For her, such initiatives do not guarantee safety and respect for trans and gender diverse people, let alone guarantee any competency around her specific needs. She suggested trans-positive messaging—such as displaying transgender pride flags—is the bare minimum that organisations should dd, but that it did not guarantee that clients could trust the organisation or feel safe there. The training and actions of staff were more important for Avery.
The participants discussed the reality of having multiple, intersecting, identities. Both Avery and Omar noted it was difficult being a “minority within a minority” (by this, Avery meant being gender non-conforming and bisexual, and Omar meant being Black and bisexual) because it meant their specific needs were often overlooked. Additionally, they noted that different dimensions of their identity meant they had different needs, obstacles, and lived experiences. For them, this highlighted the importance of housing and social service providers needing to “gain the full picture” about them and their needs. This, they observed, was difficult when there had been no trust built with providers. Organisations also did not have the resources or space to spend time—in private spaces, away from other clients—to connect with clients and learn about their specific needs.
Discussion
PV is a valuable research methodology that has the potential to give elevate marginalised voices, and is often described as being a tool for empowerment (Zoettl, 2013). It has been widely used in research with children and teenagers (Blazek & Hraňová, 2012; Haynes & Tanner, 2015; Lomax et al., 2011; Piredda et al., 2017). It has also been successfully used both in “developing” countries and with disabled communities (Huq Chávez et al., 2010; Mitchell & Lange, 2011; Sitter, 2015; Tremblay & de Oliveira Jayme, 2015). Nevertheless, when research methods do not work as planned, it is important to discuss how this reflects on the method and what can be learned from the experience. As Frazier (2020) notes, “fieldwork failure does not automatically subvert the ability to produce knowledge from the experience” (p. 142). For academic research to have rigour we must not shy away from critique.
There is limited research available regarding the failure of PV and other visual participatory methods. Much of the writing on PV, in particular, describes the benefits of the method and are thus uncritical and celebratory (Low et al., 2012). This paper serves to contribute to this much needed body of scholarship by discussing the limitations and challenges of conducting PV with a group of Takatāpui/LGBTIQ+ people who have experienced homeless. Hayward et al. (2004) note that “participatory approaches to research and development have had relatively little academic or practitioner critique, resulting in a mythologising of the power of participatory methodologies to accomplish problem solving, emancipation or empowerment” (pp. 95–96). Neglecting to discuss the challenges of participatory research methods would risk contributing to the mythologisation of such methods; hence the need for this paper. It is vital, Rogers (2016) notes, that critical scholarship of participatory methods continue to explore their complexities so that they are not taken for granted, nor assumed to be unproblematic.
Participatory research methods have often been presented as being without shortcomings in empowering marginalised peoples and creating a multi-directional relationship between researcher and participants (Hayward et al., 2004; Lenette et al., 2020; Low et al., 2012; Milne, 2016; Singh et al., 2018). However, research critiquing such methods and their ability to empower continues to emerge (Kindon, 2016a, 2016b; Lenette et al., 2020; Milne, 2016; Prins, 2010; Rice et al., 2020; Rice & Mündel, 2019; Roy et al., n. d.; Singh et al., 2018). For participants to fully engage in the research process and design, participatory research requires plenty of time, the careful building of relationships, and, often, money. A key issue that contributed to the challenges of this project is that the study population was not an existing community. Many PV projects are conducted by way of researchers going into existing communities, rather than trying to bring together a disparate group of people, as in this project. There were several constraints to this research project that meant participant and community consultation did not occur, which could have contributed to the failure of the method. For this strand of the research to have been successful, it would have been necessary to bring together these separate populations in order to find commonalities and shared goals. However, this would have taken a significant amount of time. The tight timeline of the PhD process and our desire to use multiple different research methods meant that we did not consult with these populations about how they wanted to explore the issue of Takatāpui/LGBTIQ+ homelessness prior to beginning the PV project. This was a key failing of our project; while the process of creating the videos themselves would have been participatory—had we succeeded to create any videos—the entirety of the PV project itself would have not have been fully participatory due to this lack of consultation and co-design in the planning stages.
Recruitment for the PV project was difficult because participants had stigmatised identities, and participation did not permit anonymity. Altering the method, or choosing a different participatory method, would have enabled participants to remain anonymous, but we decided against this for several reasons. Firstly, as noted in the results section, any method that required participants to come together as a group would not have enabled participants to remain fully anonymous; they would have still needed to meet with each other with the understanding that they had shared experiences of homelessness. Secondly, the tight timeline of BF’s PhD meant we ran out of time to consult with the community about an alternative, anonymous, method that we could utilise alongside the interviews already conducted. Thirdly, we wanted to adhere to the ground rules that Shaw (2012) outlined, such as all participants taking a turn speaking on camera. The PV literature emphasises the importance of all participants acting as both interviewer and interviewee throughout the PV process (Shaw, 2012; Sitter, 2012). Had we more time, community buy-in, and confidence, we would have perhaps felt comfortable to reimagine, with our participants and both homeless and Takatāpui/LGBTIQ+ communities, the PV process. We encourage future researchers who are new to the method to allow themselves the flexibility that so many argue in favour of; to allow participatory methods to shift to what the communities being researched themselves need. While the existing literature on PV is useful in providing guidance for researchers new to the method, our failed project has demonstrated the need for radical flexibility in inclusivity that allows for even “ground rules” to be dismissed if they are not serving the project.
The flexibility that would have been helpful to the research process was not available due to the limitations of a doctoral research project. The neoliberalisation of universities places significant emphasis on efficiency both of time and of finances, and encourages privatisation, competition, and self-reliance (Chávez et al., 2004). For doctoral students at our university, financial and time stressors are two of the most significant stressors placed on doctoral students (Chávez et al., 2004) Students are expected to complete in three years and scholarships only provide funding for three years, despite the average PhD taking just under 4 years to complete (Chávez et al., 2004) This is reiterated by other Aotearoa-specific research, which highlights low monetary value of postgraduate stipends and the too-short timeframe they provide students to complete their studies in (Soar et al., 2022). The need for BF to complete their PhD before funding ran out was a considerable concern, which, as touched on above, impacted on the ways in which this research was undertaken. The neoliberalisation of universities means that academics—both staff and students alike—do not always get as much time as they need to think, for ideas to ferment, and to seek out different publics and viewpoints. This has the potential to impact the quality of the academic work being produced; time and financial constraints may limit the ways in which research is conducted and shared. In the case of this research, we see that these constraints contributed to an ultimately unsuccessful PV project.
Despite BF having been deeply embedded in activism within the Takatāpui/LGBTIQ+ community for many years prior to undertaking their PhD, we still found it difficult to achieve sufficient trust within the community in the timeframe of the PV project. BF spent several years volunteering with, and ultimately running in an unpaid capacity, a local organisation that facilitated social support groups with Takatāpui/LGBTIQ+ youth. BF’s insider position as a member of the LGBTIQ+ community, combined with their years of activist experience within the community, was an asset to this research. Such a position gave BF confidence that research on Takatāpui/LGBTIQ+ people’s experiences of homelessness conducted by a member of the community was likely to be welcomed and encouraged; and indeed this was reflected by the support given to us when we recruited for interview participants. BF’s activist and personal experiences enabled them to connect with their participants and build high levels of trust during interviews. Alongside this, during their time volunteering, time they formed trusted relationships with many people within the community in Te Whanganui-a-Tara. However, when it came time to recruit for the PV project, they had not been as deeply involved in the community due to burnout, and the constant churn within voluntary organisations meant that many of the people they had working relationships with had moved on. Ultimately, this may have contributed to our lack of community uptake for the PV project; it required more vulnerability than participating in interviews, and thus more trust in the research team. Unfortunately, we were unable to build this trust in time.
Despite participants having exited homelessness, they were still dealing with the ongoing shame, stigma, and trauma of having been homeless. A key finding of BF’s PhD was the high levels of shame and stigma that participants had experiences in relation to their periods of homelessness (Fraser et al., 2021). The shame and stigma of having experienced homelessness had a direct negative impact on participants’ mental wellbeing, which contributed to the lasting trauma of having been homeless that many of them struggled with. Shame and low self-worth can result in an individualistic explanation for structural inequities, which thus positions homelessness as an individual failure (Farrugia, 2011b). Homelessness is a state of being that is heavily stigmatised as being a failure of the individual (Belcher & Deforge, 2012; Meanwell, 2012; Phelan et al., 1997; Rayburn & Guittar, 2013). We view the shame of our participants in line with Ahmed (2014), who proposes that shame can be experienced as cost of diverging from norms, and is not an individual failing. Heightened shame meant that several of the interview participants did not wish to take part in the PV project. Researchers must carefully consider if the self-disclosure inherent in such a visual methodology is too great a risk for participants who have spent much of their time and energy presenting as a different self in order to protect themselves and avoid being unnecessarily judged for their experiences of homelessness and/or their Takatapui/LGBTIQ+ identity. As Prins (2010) notes, visual methodologies are mediated by their sociocultural settings, and shame can be a significant deterrent to participation in such research. Furthermore, when participants have lived in fear, visual methodologies such as PV and photovoice may not be suitable, as they may be perceived as form of surveillance by participants (Prins, 2010). Researchers need to be aware of what they are asking from their participants and be able to provide enough protection for them. Where they cannot provide enough protection, the research methods are not appropriate.
There is little public awareness in Aotearoa of Takatāpui/LGBTIQ+ homelessness. It is understandable, then, that some people would not feel comfortable participating in a video project where they are asked to discuss two heavily stigmatised topics as they relate to their personal experiences. Much of queer activism has, and still does, focused on increasing visibility as a political tool. However, “self-disclosure is not a risk free endeavour when there is a power imbalance between the storyteller and the audience” (Holtby et al., 2015, p. 318). One of the key findings of this paper, then, is the risks of self-exposure present in this project; potential audiences would likely have had much more power, and less stigma and shame, than the participants. This risk of self-disclosure raises questions about PV’s ability to be empowering for individuals as well as communities in situations related to stigmatised topics; is it inherently empowering for an individual to disclose their potential vulnerabilities to a wide audience? Do the benefits to the wider community of study outweigh the potential detriments that participants may face in instances such as these? How can PV evolve to ensure that those with stigmatised identities/experiences are protected? These are some questions that must continue to be explored in future PV projects.
Conclusion
This paper has detailed how PV was carried out as part of a wider research project on Takatāpui/LGBTIQ+ homelessness, and specifically where it failed. Despite the lack of success in conducting a fully participatory project, this paper has shown that there are still learnings we can take from this attempt at conducting PV. Namely, it has shown how participatory methods are not always the best choice for research projects, despite their best intentions. This is particularly true when full anonymity cannot be maintained, and the subject matter involves multiple stigmatised identities and/or experiences—which can result in a lack of buy-in. This paper provides a necessary and important contribution to the existing methodological literature on PV through its critique and discussion of how the method can, and did, fail to produce the intended outcomes. For our research methods to maintain their vigour, it is important we openly discuss their failures as well as their successes. This is a necessary part of what it means to be an academic.
Supplemental Material
Supplemental Material - A Failed Attempt at Participatory Video With Takatāpui/LGBTIQ+ People Who Had Experienced Homelessness
Supplemental Material for A Failed Attempt at Participatory Video With Takatāpui/LGBTIQ+ People Who Had Experienced Homelessness by Brodie Fraser, Elinor Chisholm, and Nevil Pierse in International Journal of Qualitative Methods
Footnotes
Acknowledgments
We wish to acknowledge the research participants who generously shared with us their time and stories, as well those who did not wish to participate in this project. We respect your mana, your time, your articulate voices, and your struggles. Ngā mihi nui. Hera Cook acted as a supervisor on BF’s PhD, but did not contribute to this paper. The two no longer work together due to difference of opinion on transgender inclusion and rights. The authors wish to acknowledge transgender, non-binary, intersex, and gender expansive people’s right to self-determination.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the New Zealand Ministry of Business Innovation and Employment, Endeavour Fund (Endeavour Grant 2016).
Notes
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
